STExNMF: Spatio-Temporally Exclusive Topic Discovery for Anomalous Event Detection

Shin, Dear Sungbok; Choi, Minsuk; Choi, Jin-Ho; Langevin, Scott M.; Bethune, Christopher; Horne, Philippe; Kronenfeld, Nathan; Kannan, Ramakrishnan; Drake, Barry; Park, Haesun; Choo, Jaegul

doi:10.1109/icdm.2017.53

Cited by 12 publications

(4 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In many applications, such as text classification, information retrieval, and event discovery, feature representation is a critical issue. Traditional feature selection methods, including term frequency, MI, PLSA, and LDA, are used to generate more distinctive features [29]. However, these methods overlook the contextual information or word order in the text .…”

Section: Deep Learning Based Methodsmentioning

confidence: 99%

A Review of Research on Social Network Event Detection Methods

Chen

2024

IJCSIT

View full text Add to dashboard Cite

In recent years, with the development of technological infrastructure and the use of tech products, internet usage has become widespread globally. Significant advancements have been made in the use of social networks, which are now more readily accessible through Internet and Web 3.0 technologies such as Facebook, Twitter, and Instagram. Consequently, over the past decade, numerous researchers have been developing methods for event detection based on data collected from social media platforms. The methodologies devised for discovering events are typically modular in design and novel in terms of scale and speed. To review the research in this field, we have compiled existing works on social network event detection and conducted a comprehensive and in-depth survey. Methods for social network event detection are elaborated and categorized, with performance evaluations conducted using relevant metrics. Finally, we offer perspectives on future directions.

show abstract

Section: Deep Learning Based Methodsmentioning

confidence: 99%

A Review of Research on Social Network Event Detection Methods

Chen

2024

IJCSIT

View full text Add to dashboard Cite

show abstract

“…Each tile has time-spatial information that is extracted from tweets (i.e., posted messages) of Twitter. The authors in [17] used the spatio-temporally exclusive topic discovery based on nonnegative matrix factorization (STExNMF) method [20] to derive keywords and to conduct visualizations for detected events. Unfortunately, only about 1% of all tweets have geo-tag information [21].…”

Section: Related Workmentioning

confidence: 99%

Local Event Detection Scheme by Analyzing Relevant Documents in Social Networks

et al. 2021

View full text Add to dashboard Cite

In this paper, we propose a local event detection scheme by analyzing relevant documents in social networks to improve the accuracy of event detection. To detect local events by using geographical data, the proposed scheme embeds them using a geographical data dictionary and generates a weighted keyword graph using social network characteristics. The data left by users in social networks include not only postings but also related documents such as comments and threads. In this way, the proposed scheme detects a local event based on a keyword graph that is constructed through the analysis of the relevant documents. This can improve the accuracy of local event detection by analyzing relevant documents embedded with region-related information using a geographical data dictionary, without requiring users to tag geographic data. In order to verify the superiority of the proposed scheme, we compare it with the existing event detection schemes through various performance evaluations.

show abstract

“…Most existing methods detect topics based on a static data collection, without considering the influence of time and place on semantic. Our study is built on the online data stream, which is divided by a certain time slot and place as a basic unit like Shin et al [26]. Because the topic on the timeline is a continuous whole, with evolution, therefore, there must be some differences in addition to semantic relevance between adjacent slots.…”

Section: ) Local Topic Discoverymentioning

confidence: 99%

Burst Topic Detection in Real Time Spatial–Temporal Data Stream

Zhu

Zhang

et al. 2019

IEEE Access

View full text Add to dashboard Cite

In the field of social network, fast detection of the burst topic plays a decisive role in emergency response and disposal. However, social data are noisy and sparse, which evolves with time going on and space changing make it difficult to catch the instant semantics with traditional methods. Instead of passively waiting for an emergency topic, we try to detect the latent burst topic in its budding stage. In this paper, we propose a fast burst topic detect method, namely, FBTD, which aligns data prediction with characteristic calculation to detect burst term from the real-time spatial-temporal data stream and integrates local topic detection with global topic detection to find the spatial-temporal burst topic. Our method controls the delay within a 0.1 s level while preserving the topic quality. The experiments show that preferable effects are procured, and our method outperforms the state-of-the-art approaches in terms of effectiveness.

show abstract

STExNMF: Spatio-Temporally Exclusive Topic Discovery for Anomalous Event Detection

Cited by 12 publications

References 20 publications

A Review of Research on Social Network Event Detection Methods

A Review of Research on Social Network Event Detection Methods

Local Event Detection Scheme by Analyzing Relevant Documents in Social Networks

Burst Topic Detection in Real Time Spatial–Temporal Data Stream

Contact Info

Product

Resources

About